This is how those tweets break down, and how they look when their words are mapped to the NRC Word-Emotion Association Lexicon.
Code is available (badly commented) on GitHub, data will be available soon.
Last updated Wednesday, November 16, 2016 at 12:26 PM.
Here are some basic stats: